Background: The atopic march describes the progression from infantile food allergy (FA) to allergic rhinitis (AR) in later childhood. However, not all children with FA follow this trajectory , and predictors for AR development in this population remain poorly characterized. Objective: To identify factors influencing AR development in infants with FA and construct a predictive model for clinical application. Methods and Results: A prospective cohort included 447 infants (0-2 years) with positive food allergens, followed up to 6 years with questionnaire/clinical data. Of the 196 initially collected features, highly relevant ones were identified through ranking and comparison, and a disease prediction model was established. Post-ranking, the three key features were tobacco exposure duration, antibiotic use (notably penicillin/cephalosporin), and parental AR history , all showing positive correlations. Multinomial Naive Bayes (MNB) model (accuracy: 0.8980, F1-score: 0.8352, ROC-AUC: 0.9152) and the Multilayer Perceptron (MLP) model (accuracy: 0.9272, F1-score: 0.8791, ROC-AUC: 0.8928) demonstrated the best performance. Conclusions: The study’s machine learning prediction model has strong practical value. Early avoidance of relevant exposure factors in children at high AR risk may aid its primary and secondary prevention.